Good ways to overcomplicate your code and slow down development

Jerry Daniels jerry at daniels-mara.com
Mon Sep 18 18:03:24 EDT 2006


Shari,

I read "Getting Real" a PDF book from the folks at 37signals (they  
did Ruby on Rails and Basecamp). They have a strong philosophy  
regarding feature creep/bloat/etc. I wish I had read this earlier in  
my career. I had the same experience you did and that countless  
others have had.

Often it's better to keep a product simple and pure. If more is  
needed, it might be another product entirely with it's own source of  
funding (sales). We try and hold the line on features and  
preferences. When we make changes, we favor the type that simplify  
the product.

I also don't think it's a bad thing to start from scratch again  
sometimes. The second or third generation of development always goes  
faster and is easier to maintain. SmallTalk was re-written five  
times. In such situations, if there are customers involved, they have  
to have an easy path to cross-grade. We did that with Constellation- 
to-Galaxy. We probably should have made it 2.6 Rev compatible sooner,  
though.

Best,

Jerry Daniels

Makers of Galaxy - an alternate IDE for Revolution
http://www.daniels-mara.com

On Sep 18, 2006, at 4:31 PM, Shari wrote:

> Jerry,
>
> I wonder about this exact issue as well.  I have a program that  
> gets feature requests continually.  There is no end to the number  
> of features that can be added.  Not long ago I put out the mother  
> of all updates, adding a very long list of feature requests.  The  
> goal was to set this program aside for a good long while and work  
> on something else.  It currently has more features than any of the  
> competing programs in its price range.
>
> The logic was that if I fulfilled the most common requests and a  
> number of not so common ones, folks would be happy, more folks  
> would buy, and the "I Want" syndrome would quiet down for awhile.
>
> This did not work.  The "I Want" syndrome is alive and well, and  
> now they are just requesting more unusual features.  Sales have not  
> gone up.  They remain steady.  I expected an increase in sales for  
> the increase in features.  Sales remain the same.  Feature requests  
> still come in regularly.  So... I'm not sure there was really any  
> benefit to adding all those features.  Except for some very nice  
> reviews on the downoad sites.
>
> And the downside is... It took me away from another project for the  
> better part of a year.   A project that I expect will sell very  
> well but which is still a long time away from final release.
>
> Very frustrating!
>
> Shari
> Gypsy King Software
>
>> Josh,
>>
>> I think the earlier versions of Constellation were actually quite  
>> good, but I decided to accommodate every request (nearly) and  
>> ended up with a product that was more difficult to use than intended.
>>
>> Taking too much user feedback to heart and not sticking to my  
>> original design resulted in Constellation being very powerful for  
>> the folks who used it from its genesis, but not as easy to break  
>> into from scratch. Thus as it got more feature-laden and  
>> preference-constrained, less people adopted it, but it did have a  
>> faithful following of hundreds.
>>
>> Since Galaxy is now 2.6 Rev compatible, almost all Constellation  
>> users have moved over to Galaxy which isn't as "in your face" with  
>> its features as the latter day Constellation.
>>
>> Best,
>>
>> Jerry Daniels
>>
>> Tool makers for the 21st century
>> http://www.daniels-mara.com
>>
>
> -- 
> Gypsy King Software
> Mac and Windows shareware games
> http://www.gypsyware.com
> _______________________________________________
> use-revolution mailing list
> use-revolution at lists.runrev.com
> Please visit this url to subscribe, unsubscribe and manage your  
> subscription preferences:
> http://lists.runrev.com/mailman/listinfo/use-revolution
>




More information about the Use-livecode mailing list